Index | Search | Home | Table of Contents
Roseberg, R.J. 1996. Underexploited temperate industrial and fiber crops. p. 60-84. In: J. Janick (ed.), Progress in new crops. ASHS Press, Alexandria, VA.

Underexploited Temperate Industrial and Fiber Crops

Richard J. Roseberg

  1. MEADOWFOAM
    1. Raw Material and Products
    2. Competing Sources
    3. Crop Status
    4. Limitations
    5. Likely Commercial Production Areas
  2. FIBER FLAX
    1. Raw Material and Products
    2. Competing Sources
    3. Crop Status
    4. Limitations
    5. Likely Commercial Production Areas
  3. KENAF
    1. Raw Material and Products
    2. Competing Sources
    3. Crop Status
    4. Limitations
    5. Likely Commercial Production Areas
  4. LESQUERELLA
    1. Raw Material and Products
    2. Competing Sources
    3. Crop Status
    4. Limitations
    5. Likely Commercial Production Areas
  5. CUPHEA
    1. Raw Material and Products
    2. Competing Sources
    3. Crop Status
    4. Limitations
    5. Likely Commercial Production Areas
  6. EUPHORBIA
    1. Raw Material and Products
    2. Competing Sources
    3. Crop Status
    4. Limitations
    5. Likely Commercial Production Areas
  7. VERNONIA
    1. Raw Material and Products
    2. Competing Sources
    3. Crop Status
    4. Limitations
    5. Likely Commercial Production Areas
  8. GRINDELIA
    1. Raw Material and Products
    2. Crop Status
    3. Limitations
    4. Likely Commercial Production Areas
  9. HESPERALOE
    1. Raw Material and Products
    2. Competing Sources
    3. Crop Status
    4. Limitations
    5. Likely Commercial Production Areas
  10. HEMP
  11. SUNN HEMP
  12. HEAVY METAL HYPERACCUMULATORS
  13. OTHER POTENTIAL CROPS
  14. REFERENCES
Successful temperate new crops must either fit well into the rotation of established food, feed, or fiber crops, provide a product that is relatively more valuable than the current crop, or be better suited to growing on a given area. Proximity of growers to potential processors (who manufacture the refined product) is key to economic advantage for the new crop over prior raw material sources. In the case of arid industrial crops, the issues of survival in a harsh climate and the cost of water are paramount, while in the higher latitude temperate zones the constraints are frost tolerance, length of growing season, crop response to daylength, water requirement (rain-fed or irrigated), and weather patterns during harvest.

This paper examines the current status of several potential crops with an eye toward their requirements for success in temperate zones. Crops that are reaching or have reached active crop status include meadowfoam, fiber flax, and kenaf. Crops that are very promising, but require some further breeding, agronomy, or processing research include lesquerella, cuphea, euphorbia, and vernonia. Crops that are intriguing, but need more study to define their potential and value include grindelia and hesperaloe. Other potential crops, including those with application only in specific situations or those having received limited study and commercialization efforts are also briefly described.

The activity level for each crop is described using four new crop research and development categories put forth by L.J.M. van Soest (1993): I Plant exploration and evaluation; II Crop improvement (including plant breeding and agronomy; III Processing and application research; and IV Marketing, commercialization, and utilization. Estimating the crop's value or return to the farmer was difficult, and this difficulty increased the further away the crop was from commercialization. However, crop values were calculated using comparisons with currently available raw materials and estimates of possible differences in value.

MEADOWFOAM

The key early development for meadowfoam (Limnanthes species, Limnanthaceae) came out of the extensive USDA efforts of the late 1950s and early 1960s when many plant species were analyzed in a search for novel compounds. Out of these efforts at the National Center for Agricultural Utilization Research (NCAUR) laboratory, it was first recognized that 94% of the fatty acids in meadowfoam (Limnanthes douglasii R. Br.) seed oils had chain lengths of 20 carbon atoms or longer (Earle et al. 1959). Identification of these previously unknown long-chain fatty acids was made soon after (Smith et al. 1960; Bagby et al. 1961). Later interest shifted to Limnanthes alba Benth. due to its improved agronomic characteristics such as increased seed retention, upright growth habit, and plant height. Seed retention is no longer a problem. In fact, the cultivar Floral is considered difficult to thresh by growers. L. alba contains high levels of the same fatty acids as L. douglasii (Miller et al. 1964). Results of early agronomic and breeding work have been summarized by Jolliff et al. (1981). Results from additional crop and oil analysis were summarized by Purdy and Craig (1987).

Within the past two years the consistent involvement of industry, specifically Fanning Corp. (Chicago), has greatly stabilized the supply-demand situation, creating a steadily increasing demand. This is unlike the whipsawed supply, demand, and price history of the 1980s. Cooperation between the Oregon Meadowfoam Growers Association (OMGA), Oregon State Univ., Fanning Corp., and USDA-NCAUR has improved the coordination between crop research, product research, grower contracts, crop price, and crop area expansion.

Raw Material and Products

Meadowfoam seeds contain about 25% oil, 95% of which is made up of C:20 or C:22 monoene or diene fatty acids (Kleiman 1990). Such specificity in long chain fatty acids is rare in nature. Such fatty acids could be used in cosmetics (moisturizers, soaps, hair care), specialty lubricants and polymers (Purdy and Craig 1987; Bosisio 1989; Carlson et al. 1992).

Meadowfoam oil naturally occurs in the form of triglycerides. However, reports have described how C:40-C:44 wax esters similar to those of jojoba [Simmondsia chinensis [Link] Schneid.] and sperm whale oils could be produced from meadowfoam oil by reducing fatty acids to alcohols and then forming esters using unreacted fatty acids (Miwa and Wolff 1962; Miwa 1972; Nieschlag et al. 1977). Despite statements to the contrary (Cook 1971), meadowfoam oil is not a substitute for sperm whale oil. Unmodified meadowfoam oil is structurally quite different from sperm whale oil, and jojoba would appear to be a better source of these compounds (Kleiman 1990).

Competing Sources

Long chain fatty acids are currently produced from high erucic acid rape seed, crambe seed, and for some applications, fish oils (Jolliff et al. 1981; Purdy and Craig 1987). However, the erucic acid (C22:113) from crambe and rape seed is chemically different than the three other fatty acids (C20:15, C22:15, and C22:25,13) that make up about 85% of meadowfoam oil (Purdy and Craig 1987; Kleiman 1990).

Crop Status

Meadowfoam research and development efforts in the U.S. currently include categories I-IV. The only active crop research program is at Oregon State Univ. in cooperation with the OMGA and Fanning Corp. Product development is ongoing at the USDA-NCAUR in Peoria, Illinois, at Fanning Corp., and elsewhere. Fanning Corp. is leading market development efforts. With the current industrial demand remaining steady, the crop area is increasing from about 900 ha in 1995 to an estimated 2400 ha in 1996 in Oregon's Willamette Valley.

In Oregon, seed yields typically have ranged from 600-1200 kg/ha with current cultivars. Given the oil content and fatty acid composition, 1200 kg/ha would be worth $315/ha at rapeseed prices. However, the unique chemistry of meadowfoam oil has prompted recent contracts for $1.10/kg seed, resulting in a crop value of about $1320/ha at the high end of the normal yield range.

Limitations

The requirement for insect pollination has probably limited yields in commercial fields. Research plots can be saturated with honeybees, but in typical commercial fields only two of the five potential seeds in each flower matures. Thus, development of auto-fertile cultivars should improve yield. The range of adaptation should be further explored as new cultivars are developed. Little is currently known about range limitations.

Likely Commercial Production Areas

Meadowfoam growth on clay soils is usually acceptable, and in fact this characteristic has been one of the main reasons for its development in western Oregon. While native to western Oregon and northern California, meadowfoam should be well adapted to areas that have cool soils at planting, cool and moist weather during vegetative growth, and warm, dry harvest weather. In addition to the Willamette Valley in Oregon, such areas could include western Washington, northern Europe, New Zealand, and parts of southern Australia and southern Argentina.

FIBER FLAX

Flax [Linum usitatissimum L., Linaceae] is not a new crop. There are six references to women and flax in the Bible, indicating that flax spinning and weaving were household industries in antiquity. The center of origin of flax is thought to be in the Near East, but the exact area is a topic of debate. A highly selected cultivar dating from 5500-5000 BC was found in Iran (Dempsey 1975). Earlier flax types have been found both in Egypt and Switzerland. By 4000 BC the Egyptians had a highly developed flax industry. Other nations of the area also developed flax, and its domestication was well established in western Europe by the time of Charlemagne (742-814).

The first mechanized method for spinning flax yarn was developed in France. Between 1810 and 1820 Philippe de Girard was issued six patents relating to this new technology (Dempsey 1975). To maintain maximum fiber length and quality, flax is pulled from the ground at harvest, rather than cut. Until World War II flax was mainly pulled by hand. Since then this extremely labor intensive process has been mechanized, first by tractor-drawn machines, and then by self-propelled pullers. These machines, along with specialized turners and deseeders, were mainly developed in western Europe in the 1950s and 1960s, improving upon earlier U.S. developments during WWII and soon after. These technologies have greatly increased the speed and ease of harvest. They have also improved the utility of the inexpensive dew-retting process, whereby the pulled flax fiber is separated from cellulose and other stem parts by bacterial action in the presence of water (the dew) in the field. Field equipment and modern processing plant equipment have been further developed in western Europe during the 1980s and 1990s, allowing greater capture and separation of all classes of flax fibers while improving worker safety conditions (D. Ehrensing 1995, pers. commun.). Thus fiber recovery is currently about 25% of total biomass yield, up from about 20% forty years ago. Flax breeding was very active in the United States during the 1930s and until the 1950s, but ceased by the early 1960s (Calvert and Marks 1995). However, cultivar development has continued in Europe, especially in France, The Netherlands, and Belgium.

In the 1940s flax was grown on up to 7300 ha in Oregon (Hurst et al. 1953). The reintroduction of European flax after the end of World War II, the increase in cotton use in textiles, and the development of petroleum based fibers such as nylon combined to essentially eliminate the Oregon flax industry by the mid 1950s. Interest has recently been revived, mainly due to restrictions on stubble burning from grass seed production that have created problems for farmers in terms of weed control, insect, and disease cycles. Grass seed production occupies over 175,000 ha annually in Oregon, mainly in the Willamette Valley, (U.S. Dept. of Commerce 1993), and Oregon routinely produces over 90% of the world's perennial ryegrass (Lolium perenne L.), orchardgrass (Dactylis glomerata L.), and bentgrass (Agrostis palustris Huds.) seed. Because flax is a dicot it would provide a break in disease cycles and allow use of alternate herbicides while providing a cash crop for grass seed growers.

Raw Material and Products

Although all flax cultivars produce fiber in the stems and oil (linseed) in the seeds, fiber flax cultivars have been bred and selected specifically to produce large quantities of very long, high quality fiber, with oil production only a secondary consideration. Flax produces fibers of varying length. The longest fibers can be used in making fine linens for clothing, draperies, and furniture, medium fibers have been used for canvas and geotextiles, while short fibers have been used for paper and sacking.

Competing Sources

Fibers from oilseed flax, jute (Corchorus capsularis L.), sisal (Agave sisalana Perrine), hemp (Cannabis sativa L.), cotton (Gossypium hirsutum L.), and wood species are currently used for some of the applications suited to fiber flax. However, the high strength and quality of flax fiber makes it superior than other sources for some applications, such as linens.

Crop Status

Fiber flax research and development efforts in the United States currently include categories I-IV. In 1993 flax was grown for fiber in Russia (335,000 ha), Ukraine (127,000 ha), Belarus (120,000 ha), China (93,000 ha), and France (50,000 ha) (FAO 1994). Some fiber from oil seed flax production in Canada and northern United States (especially North Dakota) has been used recently in cigarette paper production (P.M. Carr 1995, pers. commun.).

Fiber yields in western Europe have recently been in the range of 1500-2000 kg/ha, while yields in Russia and eastern Europe have usually been less than half of those amounts (FAO 1994). This may have been due to use of oil varieties as well as poorer crop technology or management. Oregon farm fiber yields in 1995 were about 1200 kg/ha. Oregon statewide average yields from 1925 through 1951 ranged from 1.3 to 5.1 t/ha dry matter (or about 260 to 1020 kg/ha fiber) (Hurst et al. 1953).

Flax fibers that have been separated from the straw (skutched), but not combed, typically have ranged in value from $ 0.20-3.00/kg, with the longest fibers commanding the higher price (van Gelder et al. 1993; FAO 1994). Due to recent high demand, prices for short fibers have only been slightly less than those for long fibers (D. Ehrensing 1995, pers. commun.).

Limitations

Due to its long history of development, flax has few remaining problems, both in terms of agronomy and processing. The main hindrance to recommercialization in Oregon is lack of a processing plant. Estimated capital costs are $1.0 million (including the specialized field equipment) to process only short fibers, or $1.5 million to process the more valuable long fiber separately (D. Ehrensing 1995, pers. commun.).

Markets would likely expand if improved technology to use the strong flax fibers in combination with cheap, weaker fibers were developed. Flax could conceivably be mixed with excess grass seed straw or softwood fiber in composite boards or high quality papers, with cotton or polyester in clothing, or to reinforce plastics and composite materials.

Likely Commercial Production Areas

Fiber quality is enhanced by cool, moist spring weather followed by warm summers, with sufficient dew or light rain for field retting. Current production areas in northwest and eastern Europe are well-suited to fiber flax production, as are western Oregon and part of Michigan. Flax will grow in other climates, but fiber yield and quality are usually much poorer. For example, the oilseed flax production areas in north-central United States and south central Canada change from cool to hot weather rapidly, resulting in poorer fiber quality. Fiber flax production in the United States seems most likely to succeed in the Pacific Northwest due to its high fiber quality and the concentration of industry already producing fiber-based products there.

KENAF

Kenaf (Hibiscus cannabinus L., Malvaceae) has long been cultivated, probably as early as 4000 BC in Africa. Early research in the U.S. on using kenaf as a substitute for jute was begun in the 1940s due to the supply disruption from the Far East during World War II. This work continued into the 1950s, when more applied efforts, as part of the USDA Search for New Pulp Fibers program, began and continued into the 1970s (Taylor 1993). Details of this early work were well summarized by Dempsey (1975), and White et al. (1970). In 1977, the Peoria Journal Star was printed on kenaf newsprint, thus demonstrating the development of agronomic and processing technology for kenaf newsprint production. Official USDA and land grant university involvement in kenaf research was on hiatus from 1977 until 1986, when the Kenaf Demonstration Project was begun (Kugler 1988). In the meantime, several private sector efforts continued to demonstrate kenaf newsprint technology development, including the printing of several newspapers on kenaf newsprint at various times from 1977-1987 and the construction of a 200 t/day commercial pulp mill in Thailand that utilized kenaf fiber (Taylor 1993).

With the activation of the Kenaf Demonstration Project, many improvements in agronomic, processing, and newsprint production and use were made, building on the successes of the earlier efforts. An important commitment was the addition of Dr. Charles Cook to the USDA-ARS Weslaco, Texas station. Dr. Cook's primary research emphasis has been kenaf breeding and agronomy, and he has developed selections of kenaf having improved tolerance of root knot nematodes and associated pathogenic soil fungi (Cook and Mullin 1994). These pests have been significant problems in areas where kenaf was part of a cotton rotation. Development of effective harvesting, material handling, and fiber separation equipment was a breakthrough made primarily by the efforts of Harold Willett (Taylor 1993), although other separation designs also have been developed (Chen 1994). Summaries of developments since about 1987 have been published (USDA 1990; Taylor 1993: and D. Kugler elsewhere in this volume).

Raw Material and Products

The kenaf plant contains moderately long fibers in its outer stem and short fibers in its core. The outer stem (bark) makes up about 35%-40% of the stem weight, with the inner stem (core) containing the remaining 60%-65%. The fiber content of kenaf bark is about 50%-55%, increasing with plant population density, while the less valuable short fibers make up about 45%-50% of the inner core (Clark and Wolff 1969; Wood et al. 1983). Traditionally, the fiber has been used on several continents for rope, sacking, twine, and matting. However, its value will be greater if used for newsprint, carpet backing, and mixed into composite materials for boards or other structural materials (Taylor and Kugler 1992; Taylor 1993). The inner fiber has absorbent qualities that potentially could be used in products such as oil absorbents or poultry litter. Attributes of the kenaf plant were described in great detail by Dempsey (1975).

Competing Sources

For coarse fiber (lower value) applications, kenaf must compete with imported tropical monocots, chiefly jute. The higher value newsprint market in the United States is huge, and imports accounted for as much as 7.5 million tonnes, worth $4.5 billion in recent years (USDA 1993). In this market kenaf must compete with wood pulp. Increasing amounts of newsprint have been imported into the United States (up to 60% of consumption recently), mostly from Canada.

Crop Status

Kenaf research and development efforts in the United States currently include categories I-IV. Kenaf was grown on about 1660 ha in the U.S. during the early 1990s (USDA 1993). However, plans have been in place for several years to build a pulp mill requiring up to 2000 ha of kenaf mixed with recycled newspapers to produce 85 t newsprint per day (Taylor 1993). Construction has been delayed due to insufficient financing, but recent increased newsprint costs should tend to make such an enterprise more viable. The intent is still to complete the project, but details on timing and size are not yet public (C. Taylor 1996, pers. commun.). The potential area of U.S. kenaf cultivation could be as great as 0.4 to 2.0 million ha if the current upward trends in paper demand and price continue (USDA 1993; Stone 1995). Worldwide, kenaf was produced on about 200,000 ha, with major producers including China, Commonwealth of Independent States, Thailand, Cuba, India, and Mexico (FAO 1994).

Kenaf yields vary widely, not surprising given the range of areas where it has been grown and level of crop inputs. As plant density increases, stem diameters tend to decrease, but the proportion of bark tends to increase. However, the interactions between local climate, crop management, cultivar, stand density, and plant mortality make it difficult to predict stem and fiber yield without field testing (Clark and Wolff 1969; Higgins and White 1970; White et al. 1970; White et al. 1971; Dempsey 1975; Campbell and White 1982; Bhangoo et al. 1986; Scott et al. 1989). Commercial yields in the range of 9 to 22 t/ha biomass dry weight have often been reported. The higher yields were generally realized when growing conditions improved, typically as one moves from dry, high latitude locations to humid, lower latitude sites. In well adapted areas, such as the southeastern U.S., kenaf has typically yielded three to five times more fiber per year than southern pine, the typical pulping raw material source in that area (Wolff 1964; USDA 1993). Testing at several higher latitude temperate sites suggested that the adaptation of kenaf can change quite rapidly with a fairly small climatic change (White et al. 1970; Lauer 1990; Evans and Hang 1993). Results from a 1994 kenaf trial in southern Oregon showed this effect. The cultivar G4, although planted four weeks later than ideal, yielded an average 12.0 t/ha stem dry weight in the Rogue Valley, compared to 6.3 t/ha in the Willamette Valley (320 km to the north). Four cultivars that were planted four weeks before G4 yielded from 14.6 to 18.1 t/ha in the Rogue Valley.

Kenaf fiber in Thailand (where commercial kenaf pulping operations have been ongoing since 1981) were recently valued at $ 364/t (FAO 1994). To compete with wood as a pulp source in the northwestern U.S., where wood fiber is still relatively plentiful, a kenaf crop would generally sell in the range of $55-88/t (stem dry weight). For a yield of 13.5 t/ha and a price of $65/t, a grower would gross $878/ha. Part of the value of kenaf is due to the fact that pulping requires 15%-25% less energy and less chemical inputs than wood pulping (USDA 1993; Taylor and Kugler 1992).

An economic analysis for the Rio Grande Valley of Texas indicated that kenaf at $44/t would compare favorably to other crops commonly grown in that area (Scott and Taylor 1990). The overall demand for paper has continued to increase since that time. Pulp prices per tonne have rebounded from a low value of $453 in 1993 to an estimated $890 for 1995 (Brown 1995). Most analysts suggest that the dramatic price increases have been due to a combination of increased paper demand, unchanged production capacity, and increased cost and/or reduced availability of wood fiber, especially from federal timber lands in the U.S. (Hagler 1995; Stone 1995; Anon. 1995). These trends in paper demand and wood fiber availability are not expected to change. Thus, it appears that there could well be an increasing demand for quality paper fiber from non-wood sources, such as from kenaf.

Limitations

Storage and handling of annual crops like kenaf that are used in year-round manufacturing has always been somewhat of a problem. The problem is exacerbated if weather conditions or rotation requirement prevent the grower from leaving the crop standing in the field during the fall and winter after harvest. Also, while extended "vertical storage" in the field improves handling logistics, it also prevents use of the leaves as a forage (unless the plant tops were somehow harvested earlier in a separate operation). While ratios and procedures for using kenaf fiber in mixtures with other materials has been worked out in some cases (i.e. recycled newspapers for newsprint production), mixing and processing kenaf with materials such as straw, flax fiber, wood, and synthetics for composite board or other structural materials still need to be determined and optimized to broaden the applications for kenaf. Germplasm evaluation and cultivar development need to be continued, especially in terms of improved pest resistance and adaptation to higher latitudes (where summer days are longer, but seasons are shorter). Until recently, unfavorable economics had hindered the development of the kenaf industry in the United States, but research on the forage value of leaves, concurrent uses of both types of fiber produced, and increasing demand for fiber for all applications should improve the economic conditions affecting kenaf development.

Likely Commercial Production Areas

Kenaf has been grown in many areas, but highest yields have generally been observed under the following conditions: Warm soil and air (mean daily air temperatures between 22° and 30°C), sufficient moisture (monthly precipitation of 90-275 mm), fairly high relative humidity (65%-85%), a long frost free season, and fairly well drained soil which may otherwise vary greatly in texture and chemistry (Dempsey 1975). Good fertility contributes to higher yields (White et al. 1970; Dempsey 1975; Bhangoo et al. 1986). Kenaf was found to be moderately tolerant to saline irrigation water (Francois et al. 1990). Growing kenaf under irrigation modifies some of the general requirements listed above. Thus traditional growing areas in Asia, India, and the Caribbean/Central America should continue and new or increased production should be achievable in the southeast, southwest, central, and parts of the western U.S., as well as northern Australia, Africa, parts of South America, and southern Europe.

LESQUERELLA

The key early development for lesquerella (Lesquerella species, Brassicaceae) came out of the extensive USDA efforts of the late 1950s and early 1960s when many plant species were analyzed in a search for novel compounds. Out of this effort it was first recognized that Lesquerella lasiocarpa seed oils contained a high concentration of a hydroxyeicosenoic acid (Miwa et al. 1960). The structure of this C20:1-OH hydroxy fatty acid (HFA) was soon identified and given the trivial name of lesquerolic acid (Smith et al. 1961). Examination of 14 Lesquerella species showed that most had high concentrations of lesquerolic acid, while two species contained high levels of a hydroxyoctadecenoic acid (Mikolajczak et al. 1962). The structure of this C18:2-OH hydroxy fatty acid was soon identified and given the trivial name of densipolic acid (Smith et al. 1962). The relationship between native habitat and predominant HFA was recognized by Barclay et al. (1962), who noted that species native to the western U.S. usually had high concentrations of lesquerolic acid, while those native to the eastern U.S. had high concentrations of densipolic acid. However, Lesquerella auriculata, a native of Texas and Oklahoma, has seed oil with high concentrations of C20:2-OH (auricolic acid), similar in structure to densipolic acid, but having two additional carbons on the carboxyl end of the molecule (Kleiman et al. 1972). Based on morphological and agronomic characteristics, Lesquerella fendleri (Gray) Wats., a native of the southwestern U.S., was considered the most likely candidate for domestication (Barclay et al. 1962; Gentry and Barclay 1962). Early agronomic, germplasm collection, and selection studies were summarized by A.E. Thompson (1988).

Raw Material and Products

The HFAs found in high concentrations in the seed oil of several Lesquerella species could be used in producing specialty lubricants, heavy duty detergents, inks, and coatings due to the special properties of HFAs compared to other fatty acids, including higher viscosity and reactivity, caused by the presence of the hydroxyl group (Roetheli et al. 1991). Recent research has identified their utility in manufacture of plastics, additional coatings, extruded closed cell foam, facial soap, and cosmetics (Roetheli et al. 1991; Arquette and Brown 1993; Carlson et al. 1993; Thames et al. 1993a, b, 1996).

Competing Sources

Castor (Ricinus communis L.) oil contains the HFA ricinoleic acid (C18:1-OH). In recent years about 45,000 t (worth about $45 million) of castor oil are imported annually to the U.S. as its primary source of HFA. Because of this dependence on imports and the importance in certain applications, HFAs have been classified as a strategic material by the U.S. government. Because the chain length of lesquerolic acid is two carbons longer than ricinoleic acid, chemical reactions may provide more useful compounds than those from castor oil, such as the basic compound needed for nylon 1212 synthesis, whose current raw material source is petroleum (Kleiman 1990). The presence of two double bonds in densipolic and auricolic acids suggest their greater ability to undergo chemical reactions resulting in useful industrial materials. The low toxicity and low potential for irritation by lesquerella oil (unlike castor oil) suggests initial development in the high value cosmetics industry in products such as lotions, hair conditioners, and lipsticks (Arquette and Brown 1993).

Crop Status

Lesquerella research and development efforts in the U.S. currently include categories I-III. The crop has not been grown commercially, but for several years research-funded seed increases of L. fendleri have been contracted with growers in Arizona, New Mexico, and Texas, totalling 16-48 ha annually.

Based on plant morphology and limited experience with harvesting of wild stands in west Texas, Gentry and Barclay (1962) estimated seed yields of unimproved germplasm could exceed 1100 kg/ha. In more recent large field trials, seed yields in Arizona have often been between 500-1100 kg/ha, but between 1200-1800 kg/ha in small plots (Thompson et al. 1989; Thompson and Dierig 1988). In southwest Oregon, plot yields have generally been between 450-1000 kg/ha. L. fendleri seed typically contains about 25% oil of which 55%-60% is lesquerolic acid (Barclay et al. 1962; Mikolaczak et al. 1962). Castor seeds typically contain about 50% oil, of which about 85% is ricinoleic acid (Atsmon 1989). Between 1972 and 1990 castor oil prices ranged between a low of $0.51/kg in 1972 to a high of $1.60/kg in 1984, but castor oil prices have fluctuated dramatically, even within a single year (Roetheli et al. 1991). Based on these prices, calculated lesquerella seed value would be between $0.11 and $0.15/kg. If the seed yield were 1000 kg/ha, lesquerella crop value would range from $110 to 150/ha. However, lesquerella oil may be more valuable than castor oil due its applications in high value products such as cosmetics. Recent projections of seed value by potential industrial users have been in the range of $ 0.33 to 0.40/kg for improved seed containing 35% oil (Brown, 1994). Initial commercialization of lesquerella is dependent on developing such high value uses (Glaser et al. 1992; Arquette and Brown 1993).

Lesquerella fendleri is the species receiving the most research effort. Other species have been examined only at the category I and II level (exploration and crop improvement), but could also be valuable crops for several reasons. Some species produce seeds several times larger than L. fendleri (Barclay et al. 1962; Mikolajczak et al. 1962; Thompson 1988). Presumably, larger seeds would be easier to handle and possibly result in greater oil yields. Some species that produce densipolic or auricolic acid, unlike L. fendleri which produces lesquerolic acid, are native to areas outside the southwestern U.S., and thus may be adapted to production in these other locales. Some are perennials, and could be suitable for topography and cropping systems different than those that appear optimal for L. fendleri.

In two years of small plot studies in southwestern Oregon, L. grandiflora (lesquerolic acid type) plants grew well, were slightly larger than L. fendleri, and produced impressive amounts of seed. Small plots of L. perforata (densipolic acid) and L. auriculata (auricolic acid) have grown and yielded fairly well. L. purpurea and L. pinetorum (both lesquerolic acid) germinated but produced little or no seed. Plantings of L. angustifolia, L. engelmannii, L. argyraea, L. gordonii, L. gracilis, L. lasiocarpa, L. ludoviciana, (all lesquerolic acid), L. densipila, L. lyrata, and L. stonensis (all densipolic acid) failed to germinate, but the age of the seed (1972) probably had more effect than any differences in adaptation. Plantings in 1995 using several accessions (recently collected by D. Dierig) of L. douglasii (a perennial, lesquerolic acid type native to the Pacific Northwest) may further elucidate the cropping feasibility of species other than L. fendleri if grown in adapted regions.

Limitations

Because honey bees tend to very active in lesquerella plantings, it has been assumed that bee pollination enhances yield, although the degree to which insect pollination is required is not well understood. The generally lower yields seen in larger fields may be partially due to the difficulty and/or expense of achieving high bee densities and maximum seed set (Dierig and Thompson 1993). Thus, the successful domestication of lesquerella may, like meadowfoam, require the elimination of its insect-pollination requirement. Breeding work to increase oil and HFA concentrations should increase seed value.

Plant establishment has been a problem with lesquerella, due to its tendency toward seed dormancy and weak seedling vigor. Cultural practices that improve germination and stand establishment have been developed at New Mexico State Univ. (J. Fowler 1995, pers. commun.). Due to lesquerella's slow initial growth habit, competition with weeds has been a serious concern. However, several potential herbicides are good candidates for registration (Roseberg 1993; Foster et al. 1996; Roseberg 1996a), and experimental registration has been pursued in Arizona, New Mexico, and Texas (M. Foster 1995, pers. commun.).

Additional agronomic and breeding research on L. fendleri is ongoing at this time, mainly by scientists in Arizona (D. Ray, D. Dierig, J. Nelson, W. Coates, J. Brown, D. Hunsaker), New Mexico (J. Fowler), and Texas (M. Foster). Oil and gum analysis, processing, and product development research is ongoing at the USDA-NCAUR laboratory in Peoria, Illinois (T. Abbott, K. Carlson, and others), University of S. Mississippi (S. Thames), and scientists at Mycogen Corp., Lubrizol Inc., and International Flora Technologies (J. Brown and R. Kleiman).

Likely Commercial Production Areas

Lesquerella, especially L. fendleri and the other lesquerolic acid types, seem well adapted to semi-arid or arid locations. Native stands of L. fendleri are typically found in well drained, calcareous soils (Gentry and Barclay 1962). Likely production areas may include the southwest and south-central U.S., western Australia, northern Argentina and Chile, and northern Africa. L. grandiflora and L. douglasii may be adapted to parts of the western U.S. and southern Europe.

CUPHEA

The key early development of cuphea came out of the extensive USDA efforts of the late 1950s and early 1960s when many plant species were analyzed in a search for novel compounds. Out of this effort it was first recognized that seed oils from certain Cuphea species (Lythraceae) contained high levels of several medium chain triglycerides (MCTs), but that the specific dominant MCT varied with species (Earle et al. 1960; Wilson et al. 1960; Miller et al. 1964; Litchfield et al. 1967). Later studies showed that many Cuphea species contain high levels of various MCTs (Wolf et al. 1983; Graham and Kleiman 1985). An excellent summary of the genus Cuphea, including botany and early work in seed oil analyses, was written by Graham (1989).

It was reported at the First National Symposium on New Crops in 1988 that two major barriers to the domestication of cuphea were seed dormancy and seed shattering (Knapp 1990). Because seed shattering and indeterminate flowering both had been universally present in cuphea, it was impossible to harvest a significant percentage of the seed with practical field equipment. Seed dormancy has been reduced by use of C. lanceolata, whose selections contain non-dormant traits, in interspecific hybridization with C. viscosisima. The C. viscosisima phenotype exhibits desirable agronomic characteristics due to its moderate plant size, good seed size, and self-fertile behavior (eliminating the need for insect pollination) (Hirsinger 1985; Knapp 1993). As the only species native to the U.S., C. viscosisima and its phenotypes should be well adapted (Graham 1989).

The biggest barrier (seed shattering) has been partially eliminated due to the discovery, under experimental conditions, of a natural mutation within C. lanceolata x C. viscosisima hybrids that has exhibited significantly reduced shattering (Knapp 1993). The non-shattering trait has been inherited and reproduced over several generations. Additional recent key developments include the development of lines that are both non-sticky and non-shattering, as well as lines that are both auto-fertile and non-shattering (Knapp et al. 1995).

Raw Material and Products

The genus Cuphea contains about 250 wild species native to Mexico, Central, and South America (Graham 1988, 1989). Although C. viscosisima is the only species native to the U.S., four introduced species exist in the wild. A fairly large number of cuphea species have seed oils that are rich in capric, lauric, caprylic, myristic, or other medium chain fatty acids (MCFAs) (Earle et al. 1960; Miller et al. 1964; Litchfield et al. 1967; Graham and Kleiman 1985). MCFAs are used in several industrial, medical, and food applications (Bach and Babayan 1983; Arkcoll 1988). Lauric acid is used in large amounts mainly in detergent and soap products. Caprylic, capric, and myristic acid are potentially very useful MCFAs in industrial or nutritional applications (Knapp 1992).

Competing Sources

Currently the U.S. MCT demand is entirely met by imports of coconut (Cocos nucifera L.) and oil palm (Elaeis guineensis Jacq.) (Arkcoll 1988), both of which contain high levels of lauric acid. From 1986 to 1988, annual imports of palm and coconut oil were about 723,000 t to the U.S. and 995,000 t to Europe (Mackie and Calhoun 1991; Knapp 1993). The supply (and price) of tropical oils has been unstable due to variations in weather, agronomic practices, and political climate in the main producing countries while demand continues to rise (Mackie and Calhoun 1991; Carlson et al. 1992). Demand for capric acid (present in some cuphea species and suitable for plasticizer and synthetic lubricant formulations) is currently met from modified petroleum sources (Thompson 1984; Hirsinger 1985; Carlson et al. 1992).

Crop Status

Cuphea research and development efforts in the U.S. currently include categories I-III. No commercial hectarage is cultivated at this time. Breeding and agronomic work is ongoing at Oregon State Univ., with product development research at Procter & Gamble Co., Cincinnati, Ohio and the USDA-NCAUR.

Until recently the seed shattering trait of cuphea made yield estimates difficult. Using either a modified cotton picker (vacuum) for shattering types or combine for the best available non-shattering types, seed yields in Oregon often ranged from 500-1500 kg/ha. Even so, the amount of seed loss suggests that potential seed yield is well over 2000 kg/ha. The slightly warmer climate in Medford, as compared to Corvallis, Oregon, has routinely resulted in more vigorous vegetative growth for a number of Cuphea species. However, recent seed yields of the C. viscosisima x C. lanceolata hybrids have been much lower in Medford than Corvallis. The effects of moisture stress during hot weather, including reduced growth and increased seed shattering, have been observed, but yield reductions have persisted even under well-watered conditions. Perhaps the improved hybrids are better suited to a slightly cooler climate than were the wild species grown previously.

Using a coconut oil price of $453/t (FAO, 1994) and adjusting for the differences in lauric acid content (75% lauric acid in cuphea oil vs. 50% lauric acid in coconut oil) the value of cuphea oil would be $0.68/kg, or $0.24/kg seed containing 35% oil. At that price a 1500 kg/ha seed yield would be worth only $360/ha. However, as in the case of meadowfoam, the presence of unusual fatty acids (caprylic, capric, and myristic) in addition to the common (lauric), could add more value when used in new applications. This suggests that the crop should also be developed as a source of these unusual MCFAs rather than simply as a replacement for tropical oils (Knapp 1992, 1993).

Limitations

It would be desirable to combine the auto-fertile, non-sticky, and non-shattering traits into a single cultivar. Such a plant would allow a single harvest operation, possibly with a swathing operation followed several days later by threshing using a pickup-reel combine as is routinely used for grass seed harvests. The plant's stickiness can cause major equipment problems, especially if the crop is drought-stressed. However, the sticky resin may provide the plant with a protection against insects, and its removal may not be a completely positive development (Knapp 1993). As improved cultivars are developed, cultural practices must be continually evaluated. Research on product development must continue, especially in regards to the array of MCFAs available from different cuphea species.

Likely Commercial Production Areas

For the cuphea phenotypes closest to domestication, vegetative growth is favored by warm to hot weather with sufficient moisture, but seed yield is decreased under hot and dry conditions. The C. viscosisima phenotypes would likely grow well in the midwest and northwest U.S., southern Africa, southeastern South America, eastern and northern Australia, and much of Europe.

EUPHORBIA

The key early development came out of the extensive USDA efforts of the late 1950s and early 1960s when many plant species were analyzed in a search for novel compounds. Out of this effort it was first recognized that Euphorbia lagascae (Spreng.), Euphorbiaceae, was unique among the 58 euphorbs tested in that the seed oil contained high levels of vernolic (12,13 epoxy-cis-9-octadecenoic) acid. E. lagascae is a drought-tolerant native of Spain whose seed contains about 45%-50% oil, of which 60%-65% is vernolic acid (Kleiman et al. 1965; van Soest 1993; Vogel et al. 1993). Earlier, two other euphorbs (Cephalocroton cordofanus and Cephalocroton peuschelii) were shown to contain high levels of vernolic acid (Bharucha and Gunstone 1956; Gunstone and Sykes 1961). However, because they are both perennial shrubs their potential for cultivation was deemed less than that of the annual E. lagascae (Earle 1970).

The major problem with euphorbia that has hindered both breeding and agronomic research has been its violent seed shattering habit, making it difficult both to harvest and to measure seed yield. No wild accessions of euphorbia have contained a non-shattering trait (Vogel et al. 1993; Pascual-Villalobos et al. 1994). Recently, chemically induced, non-shattering mutants were developed in Spain (Pascual and Correal 1992; Pascual-Villalobos et al. 1994; Pascual-Villalobos 1996).

Raw Material and Products

Vernolic acid, the C:18 epoxy fatty acid (EFA) found in the seed oil of E. lagascae, would be useful in the paint and coating industry as a drying solvent in alkyd resin paints, a plasticizer or additive in polyvinyl chloride (PVC) resins (Riser et al. 1962; Carlson et al. 1981; Carlson and Chang 1985; Perdue 1986), and possibly in pharmaceutical applications (Ferrigni and McLaughlin 1984). Paints formulated with vernolic acid would greatly reduce volatile organic compound (VOC) air pollution that now occurs with volatilization of alkyd resins in conventional paints (Reisch 1989; Brownback and Glaser 1992; Anon. 1993). The Clean Air Act amendments of 1990 require the reduction of VOC pollutants, and regulations in California have been implemented earlier with greater effect upon the paint industry (Reisch 1989; Anon. 1993).

Competing Sources

Very few plants naturally produce high levels of vernolic acid in their seed oils (Kleiman 1990) and most of those that do have significant barriers to domestication and agronomic production (Earle 1970). The three species that appear to have the best chance for domestication include Euphorbia lagascae (Kleiman et al. 1965),Vernonia galamensis (Carlson et al. 1981; Perdue et al. 1986), and Stokes aster [Stokesia laevis Hill (Greene)] (Earle 1970; Campbell 1981).

Current sources of EFAs include epoxidized soybean oil, linseed oil (from oilseed flax), and processed petrochemicals (Carlson and Chang 1985; Perdue et al. 1986; Dierig and Thompson 1993). However, epoxidation of simple vegetable oils is an expensive process, and petrochemicals are a non-renewable and increasingly imported raw material. The U.S. currently uses about 18,000 t of EFAs in 1.2 billion liters/yr of paints and coatings alone (Reisch 1989; Anon. 1989).

Crop Status

Euphorbia research and development efforts in the U.S. currently include categories I-III, while research in Europe (especially Spain) has mainly been in categories I-II. There is no commercial hectarage at present. EFA product development is ongoing at the USDA-NCAUR and at private companies.

Four selection lines originating from the Spanish non-shattering mutants were grown for one year (1993) in Corvallis, Oregon. Seed yields in Corvallis were low, probably due to plant immaturity at the onset of cool, wet weather in the fall, but the non-shattering trait had persisted for the first generation. In 1994, four small, unreplicated, two-row plots were grown in Medford, Oregon using seed harvested in Corvallis. Plant growth was vigorous, and most plants were mature and beginning to senesce by early Oct. when they were carefully harvested by hand onto plastic tarps and dried indoors. Seed shattering was minimal, and calculated seed yields ranged from 1060 to 2800 kg/ha.

In 1995, using seed harvested in 1994, about 0.4 ha of euphorbia was grown in Medford in replicated plots including row spacing and harvest method variables. Unlike previous years, however, many plants were experiencing some degree of shattering by early Sept. Plots were either cut with an International 201 swather (draper type) or sprayed with diquat to dry the stems to allow direct combine threshing. Once dry, the euphorbia was combined in the field with a Hege 125C plot combine. Plot yields in 1995 ranged from 115-600 kg/ha. Differences were undoubtedly due to a combination of effects, including increased amount of seed shatter, use of field machinery, field drying, larger plots with minimal border effects, reduced irrigation in 1995, or other unknown factors. Based on the 1994 yields and observations of seed set in 1995, it is not unreasonable to believe that euphorbia could yield over 2000 kg/ha seed if breeding and agronomic problems were solved.

It has been observed in Europe that the percentage of vernolic acid in euphorbia seed oils varies little for different planting and harvest dates, or by differences in accession or environment (van Gelder et al. 1993; Vogel et al. 1993). This suggests that seed yield is a good predictor of fatty acid yield, and cultural practices developed to achieve maximum seed yield should also result in maximum fatty acid yield.

Average monthly prices for U.S. soybean oil ranged from $0.31 to 0.86/kg between 1982 and 1994 (Knight-Ridder Financial 1995). Based on a soybean oil price of $0.50/kg, and the value doubling or tripling after coversion to an EFA (Perdue et al. 1986), euphorbia oil would be worth $1.00 to 1.50/kg oil, or $0.50 to 0.75/kg seed. This value range was confirmed in a discussion with industry personnel. At $0.60/kg seed and 1500 kg/ha seed yield, the gross return to the farmer would be $900/ha. Of course, if euphorbia could be grown with reduced input costs, especially irrigation, its net return to the farmer would be greater than other crops having similar gross returns.

Limitations

Although the seed shattering trait has made progress on euphorbia difficult, availability of the non-shattering trait should prove to be very useful in breeding. Other characteristics present in some Spanish mutants would also increase seed yield, and could prove useful in breeding programs. Some mutants had four or five seeds per pod, instead of the three per pod observed with wild types (Pascual and Correal 1992; Vogel et al. 1993). While euphorbia's mainstem crown normally has three main terminal branches, some mutants in Spain had up to nine main terminal branches per mainstem crown, resulting in a greater number of inflorescences. Plants with such "mobheads" had shorter mainstems, possibly resulting in greater lodging resistance, and also may be more determinate in seed maturation (Pascual-Villalobos, 1996).

Breeding work has barely begun with euphorbia, but the availability of the non-shattering trait appears to be a breakthrough that can be exploited. E. lagascae is highly self-fertile, with pollen transfer occurring before insects can access the floral organs (Vogel et al. 1993). Therefore, outcrossing should be limited. Agronomic requirements of this crop have not been quantified, but now can be evaluated using non-shattering lines. Areas needing study include water and fertility requirement, planting and harvesting techniques, and seed cleaning and processing. Weed control research is also necessary, although some preliminary evaluations of potential herbicides have been made (Vogel et al. 1993). Due to the presence of latex and other potentially irritating compounds in the stems and petioles, it will be important to understand which safety precautions are necessary during harvest and processing. Processing chemistry and product development should continue on a larger scale as more seed becomes available.

Likely Commercial Production Areas

Euphorbia requires a long season for maximum seed production. It has appeared to be very tolerant to drought and heat (although the effect on yield is not known). Likely areas for production include much of the western U.S., northern Africa, southern Europe, northwest Argentina, and southern and western Australia. It should be able to compete well in situations where irrigation water is expensive or unavailable.

VERNONIA

The investigation of Vernonia anthelmintica (L.) Willd., Asteraceae, by Gunstone (1954) was the first report of a natural epoxy fatty acid in a seed oil (Earle 1970). As part of the extensive USDA efforts of the late 1950s and early 1960s to analyze many plant in search of novel compounds, it was confirmed that V. anthelmintica contained high levels of vernolic acid (Smith et al. 1959). Following that, a number of other Vernonia species were examined (Earle 1970). For a time, V. anthelmintica seemed to be the species with the greatest potential for domestication, and quite a bit of agronomic and utilization research was performed (summarized by Perdue et al. 1986). However, poor seed retention remained a major barrier to domestication. During late 1966 and early 1967, C.E. Smith Jr. collected some vernonia seeds in Africa from plants later identified as several sub-species of Vernonia galamensis [Cass.] Less. (Smith 1971). These contained about 30% oil, of which up to 78% was vernolic acid. A plant exploration trip to Africa by R.E. Perdue, Jr. in 1964 resulted in collection of seed from impressive mature plants that was later identified as a sub-species of V. galamensis different than that collected by Smith (Perdue et al. 1986). This seed contained about 42% oil, of which about 73% was vernolic acid (Carlson et al. 1981). Thus, this wild germplasm contained roughly 30% more vernolic acid than the best improved V. anthelmintica cultivars (Princen 1979). Based on the higher oil content, its observed seed retention, seed density, and seed size in the wild, it appeared V. galamensis had the greater potential for domestication.

One initial problem that discouraged domestication at temperate latitudes was short-day flowering requirement and frost intolerance. Thus, at temperate latitudes, the short days required for flowering were rapidly followed by freezing temperatures, preventing seed production (Perdue 1988; Phatak et al. 1989). A major breakthrough was the discovery that one accession of V. galamensis subsp. galamensis var. petitiana (Gilbert 1986) was day-neutral in flowering habit (Dierig and Thompson 1993) and auto-fertile. The day-neutral flowering habit has been retained by several generations of hybrids (Dierig et al. 1995). Although var. petitiana has several agronomic shortcomings, a breeding and selection program using hybrid crosses with var. ethiopica (which exhibited good plant vigor, large seed heads, large seeds, and good seed retention) has been ongoing with cooperators (including the author) in several states. Results suggest that further selection could result in vernonia hybrids that produce economically viable yields in several locations in the continental U.S. (D. Dierig 1995, pers. commun.).

Raw Material and Products

Seed oil of V. galamensis contains large amounts of vernolic acid (Perdue et al. 1986), an epoxy fatty acid identical to that found in euphorbia. However, about 45% of the vernolic acid in vernonia is in the trivernolin form, more than double that of euphorbia or stokes aster (Tallent et al. 1966; Earle 1970; Plattner et al. 1978; Carlson et al. 1981). For some applications this richer trivernolate content in vernonia may impute greater value than the vernolic acid from euphorbia or Stokes aster. In general, the product applications would be identical as those discussed for euphorbia.

Competing Sources

Competing sources for EFAs found in vernonia would be essentially the same as those listed for euphorbia.

Crop Status

Vernonia research and development efforts in the U.S. currently include categories I-III. No commercial hectarage is currently grown in the U.S. Small hectarage crops have been grown in Africa and South America using wild types to produce oil for analysis and processing research as well as improving agronomic knowledge (Anon. 1993; Carlson et al. 1992). Approximately 30% of vernonia's seed weight is vernolic acid, as in euphorbia. As such, the crop would have equivalent value, assuming extraction and processing requirements were similar. Reported seed yields in Africa using wild types have been as high as 2000 kg/ha, although details of experimental conditions have been limited. Day-neutral hybrids tested at several temperate North American sites in 1994 yielded between 166 and 1200 kg/ha (D. Dierig 1995, pers. commun.). If a seed yield of 1500 kg/ha were realized, the gross return to the farmer would be $900/ha at a value of $0.60/kg seed (the same as euphorbia). Involvement by USDA and university scientists, private industry, and the coordination of crop, processing, and product research has increased the chances for successful domestication of vernonia.

Limitations

While vernonia exhibits a dense canopy and thus competes well with weeds once it is established, early season weed control techniques will be essential during the seedling and early vegetative stage. Field tests have identified several herbicide compounds that should prove useful in vernonia cultivation (Roseberg 1996b). Seed retention and yield both need to be improved for vernonia to be commercialized. Ongoing breeding and selection programs increase the chances of improving these traits. Many questions regarding vernonia's agronomic requirements remain unanswered, including planting, water, fertilizer, and harvesting. Further evaluation and registration of herbicides or alternate weed control strategies is required. Some vernonia plants also suffer from a "sudden death syndrome," whereby an entire individual plant, at any growth stage, will wilt, senesce, and die, sometimes within a matter of hours. This has been observed at several locations in the U.S., and does not seem to be correlated to hybrid or management practice. It is assumed these are bacterial, fungal, or viral attacks, but a pathogen has not been positively identified. Worker safety during harvest and processing needs to be better understood due to the types of oils present in vernonia. Seed processing and product development research should continue in order to utilize vernonia's seed oils effectively.

Likely Commercial Production Areas

Vernonia seems to prefer well drained soil. It is also fairly drought tolerant once established, and requires a long season for maximum seed set. Production areas would likely include much of the temperate U.S., south and east Africa, parts of Argentina and southern Australia. Improved day-neutral cultivars would be required in all but the upland tropical areas of Africa and possibly parts of South America.

GRINDELIA

Several species of the genus Grindelia produce a diterpene resin (grindelic acid) and other resinous compounds on the surface of their flowers, leaves, and stems (Guerreiro et al. 1981; Bohlmann et al. 1982; Timmermann et al. 1983). Grindelia camporum (Greene), Asteraceae, is a native of the western U.S., especially the central valley of California (Bailey 1976; Hoffmann et al. 1984; Hoffmann and McLaughlin 1986). The identification of grindelic acid as an important constituent of this resinous plant (Bohlmann et al. 1982), came out of a previous examination of 195 plant species as to their suitability for agriculturally derived crude oils in the arid southwestern U.S. (McLaughlin and Hoffmann 1982), and was a key factor in the initial development of this species. Diterpene resin acids constituted between 65 and 75% of the total crude resin in the plant (Hoffmann and McLaughlin 1986; McLaughlin 1986a, b). These diterpene acids have properties similar to wood rosin and its derivatives, and form the basis for grindelia's economic crop potential (Hoffmann 1985; Hoffmann and McLaughlin 1986). Studies on the heritability of resin production indicate that genetic improvement is feasible (Dunford 1964; McLaughlin 1986a, b; McLaughlin and Linker 1987). Glands on the surface of flowers are responsible for most of the resin production, with little resin produced on the stems or leaves (Hoffmann et al. 1984).

Raw Material and Products

The diterpene resins from grindelia are very similar to those obtained by grinding up stumps or tapping live old growth pine trees (Hoffmann and McLaughlin 1986; Thompson 1990) and are known as "naval stores." They consist of a large group of compounds, including turpentine, fatty acids, rosins, and their derivatives (Thompson 1990) once used to calk ships, but now used in paper sizing processes as well as producing rubber, chemicals, ester gums, and resins for many applications (Hoffmann and McLaughlin 1986). U.S. consumption of naval stores from old growth pine trees has been fairly static at about 500,000 t/yr (Hoffmann and McLaughlin 1986), but domestic production has almost disappeared, as supplies of old growth stumpage has decreased and costs of tapping the decreasing number of live trees has increased.

Crop Status

Grindelia research and development efforts in the U.S. currently include categories I-II, with no commercial production. Grindelia camporum has received the most interest in the U.S. due to its greater biomass production compared to other resin-producing U.S. natives, such as G. squarrosa (Pursh) Dunal., G. latifolia Kellogg, and G. stricta DC. (Bailey 1976; Ravetta et al. 1995). Recently, research has begun in Argentina using G. chiloensis (Cornelissen) Cabr., a South American native that exhibited a much higher resin content on a whole-plant basis (although lower total biomass) than G. camporum (Bailey 1976; Guerreiro et al. 1981; Ravetta et al. 1995). Large numbers of resin glands are found on the stems and leaves of G. chiloensis (D. Ravetta, 1995, pers. commun.), whereas resin glands occur mainly just on inflorescences of G. camporum (Hoffmann et al. 1984; Hoffmann and McLaughlin 1986).

In Arizona, experimental tetraploid lines of G. camporum have yielded up to 12.5 t/ha biomass per year, with a crude resin content of up to 11%, resulting in annual crude resin yields of up to 1.2 t/ha, using between 57 and 75 cm of irrigation water (Hoffmann and McLaughlin 1986; McLaughlin and Linker 1987). In more recent Arizona tests, G. camporum single plant biomass yields were about twice that of G. chiloensis, although resin content was about two-thirds higher for G. chiloensis (Ravetta et al. 1995).

In southern Oregon, plantings were made in 1992 and 1993 at two locations, one a deep sandy loam and the other a moderately deep cracking clay. In 1993 plants grown in the sandy loam yielded an oven-dry biomass average of 35.2 t/ha, vs. 6.8 t/ha in the clay. Analysis of the separated flower, leaf, and stem fractions resulted in an overall average resin yield of 3.70 t/ha in the sandy loam and 0.85 t/ha in the clay. In 1994 the same plantings yielded an oven-dry biomass average of 30.4 t/ha in the sandy loam vs. 28.3 t/ha for the clay, and an overall average resin yield of 4.42 t/ha in the sandy loam and 1.66 t/ha in the clay. In these experiments, the resin yield for flowers alone ranged from 11.6% to 26.6% on an individual plant basis, with immature flowers generally having higher resin content than mature flowers. Resin content of leaves ranged from 9.2% to 21.0%, stems ranged from 1.4% to 3.6%, and combined stems with leaves ranged from 3.3% to 7.3%. Resin content was much higher in all plant parts for plants grown in the clay soil in 1993, but the opposite was true in 1994. Due to different irrigation rates, water stress was greater in the clay in 1993, but greater in the sandy loam in 1994, which may have contributed to the observed differences in yield and resin content between the two sites and years. In natural sites, grindelia's preference for soils higher in clay content (S. McLaughlin 1992, pers. commun.) may be due to the greater water holding capacity of the soil in xeric climates rather than a preference for the drainage or texture of a clay soil per se.

Grindelia seed germination is enhanced under light and is optimum at 10deg. to 20deg.C (McLaughlin and Linker 1987; R. Roseberg unpub.). Dates of transplanting and planting density effects on biomass and resin yield, crop growth rates, and water use efficiency have been examined by McLaughlin and Linker (1987). Growth and yield data from Oregon and Argentina (D. Ravetta 1995, pers. commun.) indicate that grindelia may be better adapted to dry areas that are somewhat cooler than Arizona, and that plants with high biomass yields can also have high resin content.

One factor that favors growing domesticated industrial crops in place of some traditional food crops is that their raw materials (such as resins, gums, and waxes) have typically been more valuable than food on a unit weight basis (McLaughlin 1985). Resins similar to those found in grindelia rose in price from $0.59/kg in the late 1970s to $1.10 by the mid 1980s (Hoffmann and McLaughlin 1986). While rosin prices hit a low in 1993 (Gallagher 1994a), prices have increased since then due to reduced supply and increase demand (Santos 1994; Gallagher 1994b). The increased demand for fine paper for ink-jet, laser-jet, and copier applications as well as increased use of recycled paper have increased demand for paper sizing chemicals, including rosin (Killy 1994).

If grindelia resin were valued at $0.60/kg, a biomass yield of 30 t/ha with a crude resin yield of 3.0 t/ha would be worth $ 1800/ha. Due to resin extraction and processing costs, farm gate values for raw material would be estimated at 50%, or $900/ha. Reduced input costs would contribute to a farmer's net return from a grindelia crop, mainly due its presumed reduced irrigation requirement compared to other crops. Because grindelia is a perennial, establishment costs would be reduced, further adding to a farmer's net return. Thus, the return could prove attractive in areas where other crops were limited without large inputs of expensive irrigation water.

Limitations

Resin yields need to be increased to improve grindelia's economic feasibility (Hoffmann 1985; Hoffmann and McLaughlin 1986). Genetic analysis and preliminary selection trials indicate an increase to 15%-20% resin content is possible (McLaughlin 1986a, b). Areas of highest adaptation for G. camporum need to be identified. The high concentration of resin ducts on all parts of the G. chiloensis plant suggest that this species could be domesticated in adapted areas. Also, research in Argentina with interspecific crosses of grindelia species is underway to combine the biomass production of G. camporum with the ubiquitous resin gland morphology of G. chiloensis (D. Ravetta, 1995, pers. commun.).

Many agronomic requirements of grindelia and their effects on yield and resin production are still unknown. Results in Oregon suggest the possibility of achieving high biomass production without dramatically decreasing resin content. However, the plant's resinous nature could make harvest with standard cutting equipment difficult. Processing procedures also need to be developed. Product development using grindelia biomass as a raw material also is necessary.

Likely Commercial Production Areas

Grindelia is likely suited to areas that are dry and warm. Deep soils would decrease the need for irrigation, as would presence of some clay. It is frost tolerant and a perennial, making it useful in temperate climates. Likely areas would include western U.S., western Australia, northern Africa, southern Europe, central China, and southern areas in South America.

HESPERALOE

Interest in hesperaloe (Hesperaloe funifera [Koch] Trel., Agavaceae) arose from a recent study of the fiber properties of several agaves, including hesperaloe, that are native to the southwestern U.S. and northern Mexico (McLaughlin and Schuck 1991). Hesperaloe leaves contained fiber cells that were both longer and narrower than those of sisal, and compared favorably with wood and non-wood fiber species for specialty paper applications. It had been noted that native peoples had cultivated H. funifera as a source of hard fibers (Trelease 1902; Dewey 1943) and several agavaceous species were already traded internationally due to their fiber quality, including sisal, henequen (Agave fourcroydes Lem.), and New Zealand flax (Phormium tenax J.R. & G. Forst.). It was assumed that species native to the arid southwest would be very drought tolerant. Finally, the value of pulp from hard fiber plants has typically been much higher than softwood pulps (Baker 1985; FAO 1994).

The first agronomic studies of Hesperaloe funifera were performed in Arizona. In addition to growth and yield data, preliminary information on water use, soil fertility requirements, regrowth yields and time requirements, and economic analysis were developed (McLaughlin 1993a, b, 1995). Determination that H. funifera was a crassulacean acid metabolism (CAM) plant further bolstered the observations of drought tolerance (Ravetta and McLaughlin 1993). Preliminary studies on floral biology have given insight into improved strategies for breeding and seed production (Ravetta and McLaughlin 1996).

Raw Material and Products

Hesperaloe produces hard fibers in the leaves that are very long, thin, and strong when compared with wood fibers, bast fibers, and even other hard fibers (McLaughlin and Schuck 1991; McLaughlin 1993a). Hard fibers have been successfully used to make thin, high value, specialty papers requiring strong tear resistance, such as are used in making currency, tea bags, filter papers, and specialty books (Clark 1965; Corradini 1979).

Competing Sources

Some of the world's important agricultural fibers are classified as bast fibers, obtained from phloem cells of dicot stems such as flax, kenaf, jute, ramie (Boehmeria nivea L.), and hemp. However, many important fibers are classified as hard fibers, obtained from the leaves of tropical monocots, such as abaca (Musa textilis Née), sisal, henequen, and New Zealand flax. Unlike other fibers, hesperaloe fiber is naturally white, thus eliminating the need for bleaching (Steve McLaughlin 1993, pers. commun.).

Crop Status

Hesperaloe research and development efforts in the U.S. currently include categories I-II. There is no commercial hectarage of hesperaloe. Until very recently all research plots have been small and located near Tucson, Arizona, but there are research plantings of several hectares at Maricopa, Arizona (McLaughlin 1995), and a small observation trial was planted near Medford, Oregon.

In Arizona, hesperaloe biomass accumulated to between 20 and 77 t/ha fresh weight after three years, depending on stand density (McLaughlin 1993a) and after five years of growth, yields increased to between 87 and 192 t/ha, again depending on stand density (McLaughlin 1993b). Under very good conditions, fresh weight yields in Arizona could reach about 200 t/ha (20 t/ha dry fiber) after five years (McLaughlin 1995). The growth rate accelerated after development of lateral rosettes (hesperaloe's growing point), which occurred mostly during the fourth year. Thus biomass accumulation after five years was greater than the sum of two harvests at three and five years (McLaughlin 1993b).

A small planting in southwest Oregon was started in 1992 and expanded in 1993 and 1994, to observe whether hesperaloe would survive cool, low light conditions during winter. Survival has been good for the 1993 and 1994 plants, with rodent damage during the dry summer causing more serious problems than winter weather. Growth, however, has been slow in Oregon. Yields after three years, estimated using the non-destructive technique developed by McLaughlin (1993a), were 9 t/ha (fresh weight) at densities of about 27,000 plant/ha, compared to 77 t/ha observed in Arizona. About 50% of the total biomass was added during the third year in the Arizona trial, while 90% was added during the third year in Oregon. Thus during the first and second years, survival is the main issue for hesperaloe in cooler, non-native areas. Whether or not the increased growth in subsequent years is enough to justify growing the crop will depend on relative value and return from other uses of the land.

Crop value for hesperaloe is very difficult to estimate until processing procedures are developed. However, preliminary market analysis suggested a potential processed value of $400-600/t dry fiber, corresponding to $40-60/t fresh weight (McLaughlin, 1993a). At $50/t fresh weight, a third year harvest in Arizona of 77 t/ha would be worth $3850/ha, and a fifth year harvest in Arizona of 200 t/ha would be worth $10,000 once processed. Processing costs could decrease the farm gate price to 50% of the processed fiber value, and until a market is developed, only crude estimates of crop price and value are possible.

Limitations

Although hesperaloe produces a high quality fiber, decortication of the leaves to produce dry, clean fiber has been difficult (McLaughlin 1995). There have been preliminary studies of hesperaloe genetics (Pinkava and Baker 1985; McLaughlin 1996), but the perennial habit of hesperaloe necessitates long-term efforts for breeding as well as agronomic studies. Its slow growth rate, especially in the first two years, makes weed control difficult on a large scale (S. McLaughlin 1995, pers. commun.). The range of adaptation is still unknown. The prospects and progress of hesperaloe domestication are discussed in more detail by Steven McLaughlin elsewhere in this volume.

Likely Commercial Production Areas

Hesperaloe funifera is found in the wild at low elevations in the east-central part of the Chihuahuan Desert in Coahuilla and Nuevo Leon, Mexico. H. nocturna Gentry, a species that produces similar fibers, is a native of the Sierra Madre Occidental area of northeastern Sonora, Mexico (McLaughlin 1993a; Ravetta and McLaughlin 1993). Based on research to date, hesperaloe production would seem limited to hot and dry locations near existing paper facilities, where its drought tolerance and high value fiber would make it competitive. The range of adaptation is still unknown, but its survival in southwest Oregon was somewhat surprising. Growth and yield is probably inadequate at higher latitudes, but there may be situations where it could be grown outside of Arizona and Mexico in locations less suited for conventional crops. Likely production areas include parts of the western U.S., Mexico, northern Africa, and western Australia. In any location, first year survival is the key to success, after which its perennial nature and regrowth from rosettes after cutting are advantageous.

HEMP

Hemp (Cannabis sativa L., Cannabaceae) has been grown in some parts of the world since ancient times, and it produces a high quality fiber (Dempsey 1975). Research programs have been most active recently in Europe, and production is a reality in eastern Europe and France (van Soest 1993). Restrictions on cultivation and research in the U.S. have been due to the potential for illegal drug production (marijuana). However, selections with low cannabinol levels have been developed (van Soest 1993), which may help decrease this problem. Progress in commercial production of hemp pulp for specialty papers in Europe is discussed by Anthony Capelle in more detail elsewhere in this volume.

SUNN HEMP

Sunn hemp (Crotalaria juncea L., Leguminosae) produces a bast fiber, similar to kenaf, that could be readily used in pulp and paper applications (Dempsey 1975; Cook and White 1995). Unlike kenaf, it is highly resistant to root-knot nematodes and thus can be grown in some areas where kenaf cannot (Cook and White 1995). Sunn hemp is fairly drought tolerant, can grow in marginal soils, and, being leguminous, has low nitrogen requirements. Sunn hemp is discussed by Charles Cook in more detail elsewhere in this volume.

HEAVY METAL HYPERACCUMULATORS

Various heavy metals may be present in soil naturally as well as due to mining or industrial operations, and areas of high metal concentrations occur throughout the U.S. (Holmgren et al. 1993). There are certain plants that "hyperaccumulate" various heavy metals such as Cd, Co, Cr, Cu, Mn, Ni, Pb, or Zn from the soil, taking them up in very large amounts that are unrelated to any known physiological need. Excellent summaries describing many hyperaccumulating species, representing several families, were written by Baker and Brooks (1989) and Baker et al. (1994a). A large number are in the Brassicaceae family, including Thlaspi and Alyssum species. Dr. Rufus Chaney of the USDA-ARS Environmental Chemistry Laboratory in Beltsville, MD, has examined numerous species regarding their heavy metal hyperaccumulation potential (Comis 1995). Some of these plants are very small and do not have any agronomic potential, except that this trait might be transferred to more agronomically suitable species. However, some of these plants are large enough that they aleady have potential as "biomining" crops, such as Alyssum tenium Halacsy and Dichapetalum gelonioides (Bedd.) Engl. subsp. tuberculatum Leenh. (Dichapetalaceae) for nickel, and possibly alpine pennycress (Thlaspi caerulescens, J. & C. Presl.) for Cd, Zn, and Pb (Homer et al. 1991; Baker et al. 1994a, b). Such crops would be valuable in reclaiming tailings from traditional mining operations, but also in mining the metal directly from naturally enriched soils, allowing the metal to be extracted from the crop biomass after cutting and ashing. This approach has the potential of allowing mining of heavy metals to occur in a more environmentally friendly way (Chaney 1983; Baker et al. 1994a; Comis 1995).

As an example, there are serpentine soils containing high concentrations of nickel in parts of southwestern Oregon and northwestern California, a region containing the only commercial Ni smelting operation in the U.S. (Helgerson and McNabb 1985). If one could grow a perennial bushy plant producing 40 t/ha biomass annually that contained 2% Ni, the metal yield would be 0.8 t/ha. At $7.00/kg, the refined value of the Ni would be $5600/ha (R. Chaney 1994, pers. commun.). Although processing and refining costs would be considerable (as they are in conventional mining operations), the potential to produce a crop in unproductive heavy metal-affected soil or even decontaminating mine spoils would provide a double benefit both to farmers and the environment.

OTHER POTENTIAL CROPS

There are many plants that naturally produce valuable and interesting compounds with potential industrial applications. An excellent overview of several of them was given by Robert Kleiman at the Association for the Advancement of Industrial Crops annual meeting in Catamarca, Argentina, during Sept., 1994. Those plants included: Stokes aster, Dimorphotheca pluvialis (L.) Moench; Crepis biennis and C. alpina; coriander (Coriandrum sativum L.); Calendula officinalis L.; and money plant (Lunaria annua L). Their development awaits the coordinated efforts of crop production, processing, product development, and marketing before domestication will become a reality.

REFERENCES

Last update June 3, 1997 aw